Ultra-fast interpretable machine-learning potentials

Ultra-fast interpretable machine-learning potentials
复制标题

超快速可解释的机器学习潜力

DOI:
10.1038/s41524-023-01092-7
复制
发表时间:
2023
影响因子:
9.7
通讯作者:
Hennig, Richard G.
Hennig, Richard G.
中科院分区:
材料科学1区
文献类型:
--
作者:
Xie, Stephen R.;Rupp, Matthias;Hennig, Richard G.

文献摘要

参考文献

被引文献

相似文献

全原子动力学模拟是物理、化学和材料科学中不可或缺的定量工具,但由于计算效率和预测准确性之间的权衡,大型系统和较长的模拟时间仍然具有挑战性。为了应对这一挑战,我们将三次 B 样条基础上的有效二体和三体势与正则化线性回归相结合,以获得物理上可解释的机器学习势,对于应用来说足够准确,与最快的传统经验势一样快,比最先进的机器学习势快两到四个数量级。对于来自经验势的数据,我们证明了势的精确检索。对于来自密度泛函理论的数据,预测的能量、力和导出的属性(包括声子谱、弹性常数和熔点)与参考方法的数据非常匹配。引入的势可能有助于在长时间尺度上对大型原子系统进行精确的全原子动力学模拟。
All-atom dynamics simulations are an indispensable quantitative tool in physics, chemistry, and materials science, but large systems and long simulation times remain challenging due to the trade-off between computational efficiency and predictive accuracy. To address this challenge, we combine effective two- and three-body potentials in a cubic B-spline basis with regularized linear regression to obtain machine-learning potentials that are physically interpretable, sufficiently accurate for applications, as fast as the fastest traditional empirical potentials, and two to four orders of magnitude faster than state-of-the-art machine-learning potentials. For data from empirical potentials, we demonstrate the exact retrieval of the potential. For data from density functional theory, the predicted energies, forces, and derived properties, including phonon spectra, elastic constants, and melting points, closely match those of the reference method. The introduced potentials might contribute towards accurate all-atom dynamics simulations of large atomistic systems over long-time scales.
DOI: 10.1103/physrevb.87.184115
发表时间: 2013-05-28
期刊: PHYSICAL REVIEW B
影响因子: 3.7
作者:
Bartok, Albert P.;Kondor, Risi;Csanyi, Gabor
通讯作者: Csanyi, Gabor
DOI: 10.1088/2632-2153/abf817
发表时间: 2021-09-01
影响因子: 6.8
作者:
Bircher, Martin P.;Singraber, Andreas;Dellago, Christoph
通讯作者: Dellago, Christoph
原子间势的多目标优化及其在 MgO 中的应用
DOI: --
发表时间: 2019
影响因子: 1.8
作者:
E. J. Ragasa;C J O’Brien;R G Hennig;S. Foiles;S. Phillpot
通讯作者: S. Phillpot
DOI: 10.1016/j.commatsci.2021.110752
发表时间: 2021-12
影响因子: 3.3
作者:
Joshua A Vita;D. Trinkle
通讯作者: Joshua A Vita;D. Trinkle
DOI: 10.1038/s41524-020-0283-z
发表时间: 2020-03-18
影响因子: 9.7
作者:
Vandermause, Jonathan;Torrisi, Steven B.;Kozinsky, Boris
通讯作者: Kozinsky, Boris